INDEX

Explanations

words related to layers and materials like tissues

oai_token-act-pair · gpt-3.5-turbo

terms related to layers, both physical and metaphorical, that imply complexity or concealment

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

owder

-0.63

VID

-0.62

 Defendants

-0.61

scape

-0.59

perties

-0.58

EG

-0.57

Delivery

-0.57

LET

-0.56

âĸ¬

-0.56

ECH

-0.54

POSITIVE LOGITS

peed

1.33

mith

1.29

hift

1.27

heet

1.24

pace

1.24

paces

1.22

poons

1.22

chool

1.17

creen

1.15

omething

1.13

Activations Density 0.601%

words related to layers and materials like tissues

terms related to layers, both physical and metaphorical, that imply complexity or concealment

No Comments

No Known Activations